Pep22 07 03 001

Preliminary Findings from
Drug-Related Emergency
Department Visits, 2021
Drug Abuse Warning Network

(DAWN)
Preliminary Findings from Drug-Related Emergency Department Visits, 2021:
Results from the Drug Abuse Warning Network
Acknowledgments
This report was prepared for the Substance Abuse and Mental Health Services Administration (SAMHSA), U.S.
Department of Health and Human Services (HHS), under Contract No. HHSS2832017000031I/HHSS28342002T.
Suparna Das, Rong Cai, and Karuna Nerurkar were the SAMHSA authors. Rong Cai served as the government
contracting officer representative.
Public Domain Notice

All material appearing in this report is in the public domain and may be reproduced or copied without permission
from SAMHSA. Citation of the source is appreciated. However, this publication may not be reproduced or
distributed for a fee without the specific, written authorization of the Office of Communications, SAMHSA, HHS.
Electronic Access and Printed Copies

This publication may be downloaded or ordered at https://store.samhsa.gov. Or call SAMHSA at 1-877-SAMHSA-7
(1-877-726-4727) (English and Español).
Recommended Citation
Substance Abuse and Mental Health Services Administration. (2022). Preliminary Findings from Drug-Related
Emergency Department Visits, 2021; Drug Abuse Warning Network (HHS Publication No. PEP22-07-03-001).
Rockville, MD: Center for Behavioral Health Statistics and Quality, Substance Abuse and Mental Health Services
Administration. Retrieved from https://www.samhsa.gov/data/.
Originating Office
Center for Behavioral Health Statistics and Quality, Substance Abuse and Mental Health Services Administration,
5600 Fishers Lane, Room 15-E09D, Rockville, MD 20857. For questions about this report, please e-mail
CBHSQrequest@samhsa.hhs.gov.
SAMHSA Nondiscrimination Notice

The Substance Abuse and Mental Health Services Administration (SAMHSA) complies with applicable Federal
civil rights laws and does not discriminate on the basis of race, color, national origin, age, disability, religion, or
sex (including pregnancy, sexual orientation, and gender identity). SAMHSA does not exclude people or treat
them differently because of race, color, national origin, age, disability, religion, or sex (including pregnancy, sexual
orientation, and gender identity).
U.S. Department of Health and Human Services

Substance Abuse and Mental Health Services Administration
Center for Behavioral Health Statistics and Quality
Treatment Services Branch
Publication Number PEP22-07-03-001

May 2022
DAWN Preliminary Findings from Drug-related Emergency Department Visits, 2021

Contents
1. Highlights 1
1.A Weighted National Estimates for the Top Five Drugs in Drug-Related ED Visits 1
1.B Key Findings from Sentinel Hospitals, April 2021 – November 2021 1
1.C Drugs New to DAWN’s Drug Reference Vocabulary 2
2. Introduction and Background 3
3. Methods 3
3.A Design 3
3.B Data Abstraction, Drug Reference Vocabulary, and Quality Control 4
3.C Data Notes and Limitations 5
3.D Characteristics of Unweighted Drug-Related ED Visits from DAWN Hospitals 5
3.E Weighting and Estimation 7
4. Weighted National Estimates of the Top Five Drugs in Drug-Related ED Visits, 2021 7
4.A Estimated Percentages of Alcohol-Related ED Visits 9
4.B Estimated Percentages of Opioid-Related ED Visits 11
4.C Estimated Percentages of Methamphetamine-Related ED Visits 15
4.D Estimated Percentages of Marijuana-Related ED Visits 17
4.E Estimated Percentages of Cocaine-Related ED Visits 19
5. Key Findings from Sentinel Hospitals, April 2021 – November 2021 21

5.A Sentinel Hospitals: Top Ten Drugs in Drug-Related ED Visits 21
5.B Sentinel Hospitals: Polysubstance ED Visits 22
6. Drugs New to DAWN Drug Reference Vocabulary, 2021 24
APPENDICES
Appendix A. Methods 25
Appendix B. Definitions of the Top Five Drugs in Drug-Related ED Visits 31
Appendix C. Data Tables, Top Five Drugs in Drug-Related ED Visits, 2021 32
DAWN Preliminary Findings from Drug-related Emergency Department Visits, 2021 i

TABLES
Table 3.A.1 Overview of the DAWN hospital selection 4

Table 3.D.1 Demo-geographic distribution of DAWN cases (n = 141,529) 6
Table 5.A.1 Percentage of top ten drug-related visits to all drug-related ED visits by month 21
Table 6.1 Drugs new to the DAWN DRV, 2021 24
Table A1 Overview of the DAWN sample 26
Table A2 Initial base weights for Part B and Part C hospitals 29
Table A3 Adjusted base weights for Part B and Part C hospitals 30
Table B1 Top five drug definitions 31
Table C1 Estimated percentages and counts of drug-related ED visits by top five drugs, 2021 32
Table C2 Estimated subgroup percentages of alcohol-related ED visits (N = 2,942,609) 33
Table C3 Estimated subgroup percentages of fentanyl, heroin, and other opioid-related ED visits 34
Table C4 Estimated subgroup percentages of methamphetamine-related ED visits (N = 797,936) 35
Table C5 Estimated subgroup percentages of marijuana-related ED visits (N = 787,769) 36
Table C6 Estimated subgroup percentages of cocaine-related ED visits (N = 336,516) 37
DAWN Preliminary Findings from Drug-related Emergency Department Visits, 2021 ii

FIGURES
Figure 4.1 Estimated percentages and counts of drug-related ED visits by the top five drugs
(January 2021-December 2021) 8
Figure 4.A.1 Alcohol-related ED visits by age group 9
Figure 4.A.2 Alcohol-related ED visits by age (< 21 and 21+) 9
Figure 4.A.3 Alcohol-related ED visits by sex 9
Figure 4.A.4 Alcohol-related ED visits by race 9
Figure 4.A.5 Alcohol-related ED visits by ethnicity 10
Figure 4.A.6 Alcohol-related ED visits by census region 10
Figure 4.A.7 Alcohol-related ED visits by quarter 10
Figure 4.B.1 Fentanyl, heroin, and other opioid-related ED visits by age group 11
Figure 4.B.2 Fentanyl, heroin, and other opioid-related ED visits by sex 11
Figure 4.B.3 Fentanyl, heroin, and other opioid-related ED visits by race 12
Figure 4.B.4 Fentanyl, heroin, and other opioid-related ED visits by ethnicity 12
Figure 4.B.5 Fentanyl, heroin, and other opioid-related ED visits by census region 13
Figure 4.B.6 Fentanyl, heroin, and other opioid-related ED visits by quarter 13
Figure 4.C.1 Methamphetamine-related ED visits by age group 15
Figure 4.C.2 Methamphetamine-related ED visits by sex 15
Figure 4.C.3 Methamphetamine-related ED visits by race 15
Figure 4.C.4 Methamphetamine-related ED visits by ethnicity 15
Figure 4.C.5 Methamphetamine-related ED visits by census region 16
Figure 4.C.6 Methamphetamine-related ED visits by quarter 16
Figure 4.D.1 Marijuana-related ED visits by age group 17
Figure 4.D.2 Marijuana-related ED visits by sex 17
Figure 4.D.3 Marijuana-related ED visits by race 17
Figure 4.D.4 Marijuana-related ED visits by ethnicity 17
Figure 4.D.5 Marijuana-related ED visits by census region 18
Figure 4.D.6 Marijuana-related ED visits by quarter 18
Figure 4.E.1 Cocaine-related ED visits by age group 19
Figure 4.E.2 Cocaine-related ED visits by sex 19
Figure 4.E.3 Cocaine-related ED visits by race 19
Figure 4.E.4 Cocaine-related ED visits by ethnicity 19
Figure 4.E.5 Cocaine-related ED visits by census region 20
Figure 4.E.6 Cocaine-related ED visits by quarter 20
Figure 5.B.1 Polysubstance ED visits 22
Figure 5.B.2 Top five drugs involved in alcohol-related polysubstance ED visits (N = 5,353) 23
Figure 5.B.3 Top five drugs involved in methamphetamine-related polysubstance ED visits (N = 4,027) 23
Figure 5.B.4 Top five drugs involved in marijuana-related polysubstance ED visits (N = 3,106) 23
Figure 5.B.5 Top five drugs involved in cocaine-related polysubstance ED visits (N = 2,722) 23
Figure 5.B.6 Top five drugs involved in heroin-related polysubstance ED visits (N = 1,560) 23
Figure 5.B.7 Top five drugs involved in fentanyl-related polysubstance ED visits (N = 1,539) 23
Figure A1 DAWN hospital selection design 27
DAWN Preliminary Findings from Drug-related Emergency Department Visits, 2021 iii
1. Highlights
In 2021, the Drug Abuse Warning Network (DAWN) identified 141,529 (unweighted) drug-related emergency department
(ED) visits from 52 participating hospitals. These data were analyzed to (1) generate nationally representative weighted
estimates for the top five drugs in drug-related ED visits, (2) assess monthly trends and drugs involved in polysubstance
ED visits in a subset of sentinel hospitals, and (3) identify drugs new to DAWN’s Drug Reference Vocabulary (DRV) in 2021.
1.A Weighted National Estimates for the Top Five Drugs in Drug-Related ED Visits
KEY FINDINGS BY DRUGS
› The top five drugs involved in drug-related ED visits in 2021 were alcohol (39.33% of all drug-related ED visits), opioids
(14.07%), methamphetamine (11.02%), marijuana (10.78%), and cocaine (4.71%).
› Fentanyl-related ED visits rose throughout 2021, peaking in quarter 4. Heroin-related ED visits rose from quarter 1
through quarter 3, and declined in quarter 4. All other top drugs peaked in quarter 2, and declined in quarters 3 and 4.
KEY FINDINGS BY AGE, SEX, RACE, ETHNICITY, AND CENSUS REGION
› Patients aged 18 to 25 had the second highest percentage of ED visits related to marijuana (26.80%) and fentanyl
(16.90%).
› Males had a higher percentage of ED visits for all top five drugs as compared to females, however, the difference was
much smaller (54.04% males vs. 45.94% females) for ED visits related to other opioid pain medications and their
combinations.
› Black or African American patients had the highest percentage of ED visits related to cocaine (44.19%), and the second
highest percentage of ED visits related to marijuana (23.87%) and heroin (20.35%).
› More than one-third of ED visits related to fentanyl were from the West census region alone (39.65%). The West and
South census regions combined accounted for more than two-thirds of ED visits related to methamphetamine (71.02%).
The West and South regions combined accounted for the majority of visits related to other opioid pain medications and
their combinations (59.51%).
1.B Key Findings from Sentinel Hospitals, April 2021 – November 2021
› Analysis of unweighted data from DAWN sentinel hospitals from April to November 2021 identified alcohol,
methamphetamine, marijuana, cocaine, and heroin as the top five drugs involved in drug-related ED visits. Alcohol
accounted for more than half of these drug-related ED visits.
› Monthly trend analysis revealed decreasing trends of alcohol, methamphetamine, marijuana, and heroin-related ED
visits, and increasing trends of fentanyl and unspecified narcotic analgesics.
› The top six drugs involved in polysubstance ED visits (i.e., visits related to more than one drug) in sentinel hospitals
were alcohol, methamphetamine, marijuana, cocaine, heroin, and fentanyl. While the majority of alcohol-related ED
visits were due to alcohol alone, a significant percentage of methamphetamine-, marijuana-, cocaine-, heroin-, and
fentanyl-related ED visits involved at least one other drug.
DAWN Preliminary Findings from Drug-related Emergency Department Visits, 2021 1

› Alcohol was the most common additional drug involved in methamphetamine-, marijuana-, and cocaine-related
polysubstance ED visits, while methamphetamine was the most common additional drug involved in heroin- and
fentanyl-related polysubstance ED visits.
1.C Drugs New to DAWN’s Drug Reference Vocabulary
› In-depth analysis of 2021 data identified 38 substances related to ED visits that were new to DAWN’s DRV. All of these
substances were added to the DRV’s existing drug categories. Of the 38 substances, roughly one-third were illicit drugs,
and the other two-thirds were non-illicit substances.

2. Introduction and Background
Authorized by the 21st Century Cures Act in December 2016, DAWN is a nationwide public health surveillance system
administered by the Substance Abuse and Mental Health Services Administration’s (SAMHSA’s) Center for Behavioral
Health Statistics and Quality (CBHSQ). From 1992 through 2011 SAMHSA administered the former DAWN system,
referred to as legacy DAWN, which collected data from ED visits resulting from substance use and overdoses. In 2018,
SAMHSA re-established DAWN as a nationwide public health surveillance system that improves ED monitoring of
substance use-related visits, including those related to opioids. DAWN captures data on ED visits related to recent
substance use and misuse directly from the electronic health records (EHR) of participating hospitals. It helps SAMHSA
and public health professionals, clinicians, and policymakers respond effectively to the opioid and substance misuse crisis
in the United States.
The key objectives of DAWN are to:
› Monitor demographic and geographic distribution, and identify trends of substance-related ED visits;
› Provide an early warning system that identifies emerging and novel psychoactive substances and/or combinations
of substances; and
› Provide national estimates of substance-related ED visits to key stakeholders and the public.
3. Methods
3.A Design
DAWN is comprised of eligible hospitals selected from the 2016 American Hospital Association (AHA) file. All hospitals
located in the United States and the District of Columbia that are non-federal, short-stay, general surgical and medical
hospitals, with at least one 24-hour ED, and at least 100 annual ED visits, were eligible for selection (N = 4,310). DAWN
uses a hybrid design of sentinel hospital-based surveillance and probability sample-based surveillance to better assess the
epidemiology of substance use- and misuse-related ED visits.
Hospitals included in the sentinel-based surveillance were selected by SAMHSA from key urban areas with the goal
of serving as an early warning system (Part A). Sentinel hospitals are expected to yield cases of concern, allowing
for the collection and analysis of data quickly so a plan of action for addressing key and emerging problems can be
developed. Sentinel hospitals were selected from counties that had the highest counts and rates for opioids, cocaine, and
psychostimulant-related overdose deaths and ED visits.
Hospitals from the probability sample-based surveillance were selected using a stratified random sampling method and
were divided into two categories: (1) a probability sample of hospitals in high-priority suburban and rural areas (Part B) to
ensure the representation of suburban and rural hospitals, and (2) a probability sample of all the remaining hospitals (Part
C). Part B hospitals were selected from suburban and rural counties with a high rate of drug-induced overdose deaths. Part
C hospitals were selected from the remaining list of hospitals not included in Part B.

Table 1 provides an overview of the DAWN hospitals. Originally, DAWN was intended to be comprised of 50 hospitals (10
Part A, 8 Part B, and 32 Part C). The DAWN redesign in 2019 led to the retention of three hospitals, totaling 53 hospitals
for DAWN. For weighting and estimation purposes, the three additional hospitals are included in Part A. See Appendix A
for more information on the DAWN hospital selection design.
Table 3.A.1 Overview of the DAWN hospital selection
Part Number Description

Part A 13 Consists of 10 sentinel hospitals and 3 hospitals recruited prior to redesign
Part B 8 Sample of hospitals from suburban and rural counties with a high rate of drug-induced deaths
Part C 32 Sample of hospitals from all remaining hospitals not included in Part B
Participating DAWN hospitals complete a Data Access Agreement (DAA), Business Associate Agreement (BAA), and
Institutional Review Board (IRB) approvals, as required. DAWN is committed by law and by practice to preserving patient
privacy and safety by following the Health Insurance Portability and Accountability Act (HIPAA) (PL 104-191) and the HIPAA
Privacy Rule (45 CFR Part 160 and Part 164 subparts A and E). The identity of DAWN participating hospitals is confidential.
3.B Data Abstraction, Drug Reference Vocabulary, and Quality Control
DAWN’s data abstraction effort involves the direct record review of all ED visit records from participating hospitals.
Trained medical record abstractors review key areas of each patient’s ED visit record to assess whether the visit is related
to recent drug use. If it is, they abstract key data items from the record into a web-based reporting system. DAWN
abstracts demographic data, characteristics of the ED visit (time, date, etc.), and the drugs related to the visit. DAWN data
elements do not include direct patient identifiers.
DAWN defines drug-related ED visits as drug use indicated as the reason for the ED visit, either as the direct cause (e.g.,
overdose) or as a contributing factor (e.g., injury, infection, organ damage). ED visits where the patient left the ED without
being seen by a clinician or visits where the patient came to the ED but was directly admitted to the hospital are ineligible
for review. See Appendix A for more information on data abstraction.
DAWN abstractors select the drug(s) related to the visit using a menu of drugs within the web-based reporting system.
This menu is based on the contents of the DAWN DRV, which is a comprehensive drug vocabulary and classification
system for all substances collected by DAWN. The DAWN DRV is derived from the Multum Lexicon©, 2020 Cerner
Multum Inc., with modifications to meet DAWN’s unique requirements. The DRV includes codes for brand (trade) names,
generic names, chemical names, metabolites, nonspecific drug terms, and street and slang terms for legal and illegal
substances, including prescription and over-the-counter pharmaceuticals, alcohol, and select non-pharmaceuticals that
are abused by inhalation.
Each drug in the DRV is assigned a drug ID. Brand names or street terms reported are classified to the drug ID of their
more generic drug name. For example, smack, a street term for heroin, is linked to the drug ID for heroin. Drug IDs are
used in the analysis of DAWN data either individually or in defined analytic groups. Drug IDs are assigned to individual
drugs or drug combinations. When a drug-related visit is related to two or more drugs mixed together, this is defined as a

drug combination and is included on the DRV, for example, cocaine/marijuana is a mixture of cocaine and marijuana and
has its own unique drug ID. This is different than when two drugs are related to a visit but taken separately. For this report,
drug IDs were used to create a defined analytic group or the analysis was done at the drug ID level with no grouping.
DAWN uses a multi-step process to ensure data quality. Automated checks are embedded in the web-based abstraction
system at initial entry to minimize errors, and machine learning (ML) algorithms are applied to each drug-related visit
once it is submitted. Cases flagged by the ML model are subject to human review, which can be reviewed, edited, and
resubmitted by a data manager. Quality control audits are performed on each abstractor’s data bi-annually to identify
training needs.
3.C Data Notes and Limitations
The recruitment of DAWN hospitals was ongoing in 2021. During 2021, 52 hospitals were actively participating and
submitting data. Estimates presented in this report are subject to change as DAWN continues to collect data from
additional hospitals. DAWN collects data on ED visits and not individuals. This means that a patient who has multiple
drug-related ED visits will be counted as a separate drug-related ED visit each time.
The weighted estimates presented in this report are based on 52 hospitals, 13 selected for surveillance and 39 via a
probability sample. Caution should be exercised when interpreting findings, as the precision offered by DAWN’s small
sample size is limited, and can lead to unreliable estimates. DAWN used relative standard error (RSE), calculated
by dividing the standard error of the estimate with the estimate itself, as a measure of reliability. To avoid erroneous
conclusions based on unreliable estimates, estimates with RSE > 50 percent are suppressed in this report. Estimates based
on counts larger than zero and less than ten are suppressed to ensure patient confidentiality and minimize disclosure risk.
Definitions of the drugs and/or categories presented in the report are derived from the DAWN DRV and are provided in
the respective sections and/or as footnotes.
3.D Characteristics of Unweighted Drug-Related ED Visits from DAWN Hospitals
Table 2 presents the demo-geographic distribution of drug-related ED visits from participating hospitals based on
unweighted data. In 2021, 141,529 drug-related ED visits were identified from 52 hospitals. Drug-related ED visits from
participating hospitals were predominantly among patients with the following characteristics: aged 26-44 (42.76%), male
(65.65%), White (56.01%), Not Hispanic or Latino (78.40%), and from the West census region (30.61%).

Table 3.D.1 Demo-geographic distribution of DAWN cases (n = 141,529)
Subgroups Count Percent

Age Groups
Under 18 4,953 3.50
18 to 25 16,691 11.79
26 to 44 60,523 42.76
45 to 64 45,111 31.87
65 and over 13,692 9.67
Missing 559 0.39
Age Groups (< 21 and 21+)
Under 21 9,691 6.85
21 and over 131,279 92.76
Missing 559 0.40
Sex
Male 92,920 65.65
Female 48,381 34.18
Other 119 0.08
Not documented 109 0.08
Race
White 79,269 56.01
Black or African-American 33,430 23.62
Asian 3,077 2.17
American Indian/Alaska Native 4,073 2.88
Native Hawaiian/Pacific Islander 1,493 1.05
Other race 13,342 9.43
Multi-Racial* 2,735 1.93
Missing/Not documented 4,110 2.90
Ethnicity
Not Hispanic or Latino 110,963 78.40
Hispanic or Latino 19,820 14.00
Unknown 10,746 7.59
Region
Midwest 31,904 22.54
Northeast 23,338 16.49
South 24,768 17.50
West 43,324 30.61
No address/Institution 16,299 11.52
Missing 1,896 1.34
* ED visits with multiple-race categories are counted in the Multi-Racial category only.

3.E Weighting and Estimation
DAWN employed a multi-step weighting process to produce nationally representative estimates given DAWN’s hybrid
sentinel surveillance and probability sample design. The multi-step weighting process involved (1) calculating initial
base weights for each sampling part/stratum, (2) adjusting the initial base weights for changes in the sample design and
sampling frame, (3) adjusting for hospital non-response, and (4) post-stratification to adjust DAWN estimates of ED visit
totals to AHA ED visits for the given stratum. For more details on weighting and estimation, see Appendix A.
4. Weighted National Estimates of the Top Five Drugs in Drug-

Related ED Visits, 2021
This section presents weighted national estimates of drug-related ED visits. The top five drugs in drug-related ED visits in
2021 were alcohol, opioids, methamphetamine, marijuana, and cocaine. The opioids category is further broken down into
ED visits related to fentanyl, heroin, and other opioid pain medications. The top five drug-related ED visits include the
drug used alone or in combination with other drugs as follows:
› Alcohol-related ED visits involving alcohol alone or in combination with other drugs.
› Opioid-related ED visits involving fentanyl, heroin, and other opioid pain medications taken alone or in combination
with other opioids and/or other drugs.
› Methamphetamine-related ED visits involving methamphetamine alone or in combination with other drugs.
› Marijuana-related ED visits involving marijuana alone or in combination with other drugs.
› Cocaine-related ED visits involving cocaine alone or in combination with other drugs.
Analytic grouping by drug ID was used for this analysis. See Appendix B for a list of drugs and drug combinations
included.
Since a significant number of ED visits involve multiple substances, caution should be exercised while adding and/or
comparing estimates across drugs, as ED visits related to more than one drug are counted more than once. For example,
an ED visit related to both cocaine and marijuana are counted in both cocaine-related and marijuana-related ED visits.

In 2021, the majority of drug-related ED visits involved alcohol (39.33%), followed by opioids (14.07%),
methamphetamine (11.02%), marijuana (10.78%), and cocaine (4.71%). Among the opioids, the majority of ED
visits were from the heroin subcategory (6.94%) followed by other opioids and fentanyl.
Figure 4.1 Estimated percentages and counts of drug-related ED visits by the top five drugs
(January 2021–December 2021)
Estimated percentages and counts of opioid-related ED visits, 2021
15%
10%
50%
5%
40%
0%
Heroin Other opioids Fentanyl
(N = 493,916) (N = 484,735) (N = 122,884)
30%
20%
10%
0%
Alcohol Opioids Methamphetamine Marijuana Cocaine
(N = 2,942,609) (N = 1,032,510) (N = 797,936) (N = 787,769) (N = 336,516)
The top five drugs in all drug-related ED visits in 2021 were alcohol, opioids, methamphetamine,
marijuana, and cocaine. Alcohol accounted for the majority of drug-related ED visits at 39.33
percent (2.9 million), followed by opioids at 14.07 percent (1.03 million). Opioids were further
broken down into three categories—fentanyl, heroin, and other opioids. Among these opioid
categories, heroin accounted for the highest percentage.
The following sections present subgroup analyses for each of the top five drugs: alcohol, opioids, methamphetamine,
marijuana, and cocaine. Each section provides estimated percentages by age, sex, race, ethnicity, census region, in that
order, and quarterly trends in 2021. For alcohol-related ED visits, an additional figure of estimated percentages for
underage (under 21 years) vs. adult (21 years and older) is included. See Appendix C for accompanying tables.

4.A Estimated Percentages of Alcohol-Related ED Visits
(Estimated weighted count = 2,942,609)
Figure 4.A.1 Alcohol-related ED Visits by Age Group Figure 4.A.2 Alcohol-related ED Visits by Age (< 21 and 21+)
100% 100%
80% 80%
60% 60%
40% 40%
20% 20%
0% 0%
Under 18 18 to 25 26 to 44 45 to 64 65 and over Missing* Under 21 21 and Over Missing*
* Suppressed due to a relative standard error (RSE) > 0.5 * Suppressed due to a relative standard error (RSE) > 0.5
The percentage of alcohol-related ED visits was highest among Patients ages 21 and over accounted for 95.58 percent of all
patients ages 26 to 44 (40.63%) followed closely by patients alcohol-related ED visits.
ages 45 to 64 (38.92%).
Figure 4.A.3 Alcohol-related ED visits by sex Figure 4.A.4 Alcohol-related ED visits by race
80% 80%
60% 60%
40% 40%
20% 20%
0% 0%
Male Female Not documented Other* White Other Black Asian Multi-Racial NH/PI AI/AN* Missing/ND*
* Suppressed due to a relative standard error (RSE) > 0.5 *Suppressed due to a relative standard error (RSE) > 0.5
Male patients accounted for 71.01 percent of alcohol-related White patients accounted for the highest percentage of alcohol-
ED visits, while female patients accounted for 28.94 percent related ED visits (59.76%), while Other patients accounted for
of these visits. the second highest percentage of these visits (15.09%).
Note: Other—The race documented in the medical record does not fit any
other race category. Multi-Racial—Multiple races were selected. AI/AN—
American Indian or Alaska Native. NH/PI—Native Hawaiian or Other Pacific
Islander. Missing/ND—Race was missing or was not documented in the
medical record.

Estimated Percentages of Alcohol-Related ED Visits (continued)
(Estimated weighted count = 2,942,609)
Figure 4.A.5 Alcohol-related ED Visits by Ethnicity Figure 4.A.6 Alcohol-related ED Visits by Census Region
100% 100%
80% 80%
60% 60%
40% 40%
20% 20%
0% 0%
Not Hispanic or Latino Hispanic or Latino Unknown West Northeast Midwest South No address/Inst Missing
Not Hispanic or Latino patients accounted for 74.65 percent The regional percentage of alcohol-related ED visits was
of alcohol-related ED visits, while Hispanic or Latino patients highest among patients residing in the West (27.45%), followed
accounted for 14.26 percent of these visits. by patients residing in the Northeast (23.42%) and Midwest
(22.40%).
Note: No address/Inst—Patient had no fixed address or currently
resides in an institution (e.g., shelter, jail, hospital).
Figure 4.A.7 Alcohol-related ED visits by quarter

Key Takeaways from Alcohol-Related
80% ED Visits
Alcohol-related ED visits in 2021 were more likely
60%
to be Not Hispanic or Latino, White, male, among
those aged 26 to 64, and from the West census
region. When comparing underage (under 21) vs.
40%
adult (21 and over) age groups, the majority of these
patients were over 21 years of age (95.58%). The
20% South census region had the lowest percentage of
alcohol-related ED visits as compared to other census
0% regions. Alcohol-related ED visits peaked in quarter
Qtr1 Qtr2 Qtr3 Qtr4
2 before declining in quarters 3 and 4. For the data
table accompanying the figures in this section, see
The percentage of alcohol-related ED visits rose from quarter 1 to Appendix Table C2.
quarter 2 before declining into quarters 3 and 4.

4.B Estimated Percentages of Opioid-Related ED Visits
(Estimated weighted counts: Fentanyl = 122,884, Heroin = 493,916, Other Opioids = 484,735)
Figure 4.B.1 Fentanyl, Heroin, and Other Opioid-related ED Visits by Age Groups
Under 181 18 to 25 26 to 44 45 to 64 65 and over Missing2
80%
60%
40%
20%
0%
Fentanyl Heroin Other Opioids
1
The percent estimate for the Under 18 age group was suppressed for fentanyl due to a RSE > 0.5
2
The percent estimates for the Missing age group were suppressed for each opioid category due to a RSE > 0.5
Note: For the data table accompanying this figure, see Appendix Table C3.
The percentage of fentanyl, heroin, and other opioid-related ED visits was highest among patients ages
26 to 44. Patients ages 45 to 64 had the second highest percentage of heroin and other opioid-related ED
visits, while patients ages 18 to 25 had the second highest percentage of fentanyl-related ED visits.
Figure 4.B.2 Fentanyl, Heroin, and Other Opioid-related ED Visits by Sex
Male Female Other* Not documented*
80%
60%
40%
20%
0%
* The percent estimates for Other and Not documented were based on less than 10 cases in every opioid category and
therefore are suppressed
Male patients accounted for the highest percentage of fentanyl, heroin, and other opioid-related ED visits,
although the difference between males and females was smaller for other opioid-related ED visits.

Estimated Percentages of Opioid-Related ED Visits (continued)
Figure 4.B.3 Fentanyl, heroin, and other opioid-related ED visits by race
White Black Other Asian1 Multi-Racial2 AI/AN NH/PI3 Missing/ND4
100%
80%
60%
40%
20%
0%
1
The percent estimate for Asians was suppressed for heroin due to a relative standard error (RSE) > 0.5
2
The percent estimate for Multi-Racial was suppressed for heroin and other opioids due to a relative standard error (RSE) > 0.5
3
The percent estimate for NH/PI was suppressed for fentanyl and heroin due to a relative standard error (RSE) > 0.5
4
The percent estimates for Missing/ND were suppressed for every opioid category due to a relative standard error (RSE) > 0.5
Note: Other—The race documented in the medical record does not fit any other race category. Multi-Racial—Multiple races were
selected. AI/AN—American Indian or Alaska Native. NH/PI—Native Hawaiian or Other Pacific Islander. Missing/ND—Race was
missing or was not documented in the medical record. For the data table accompanying this figure, see Appendix Table C3.
White patients accounted for the highest percentage of fentanyl, heroin, and other opioid-related ED visits, while
Black patients accounted for the second highest percentage of heroin and other opioid-related ED visits.
Figure 4.B.4 Fentanyl, heroin, and other opioid-related ED visits by ethnicity
Not Hispanic or Latino Hispanic or Latino Unknown*
100%
80%
60%
40%
20%
0%
* The percent estimate for unknown ethnicity was suppressed for other opioids due to a relative standard error (RSE) > 0.5
Not Hispanic or Latino patients accounted for the highest percentage of fentanyl, heroin, and other opioid-
related ED visits.

Figure 4.B.5 Fentanyl, heroin, and other opioid-related ED visits by census region
West South* Northeast* Midwest* No address/Inst Missing
60%
40%
20%
0%
* The percent estimates for the South, Northeast, and Midwest regions were suppressed for heroin due to a relative standard
error (RSE) > 0.5
Note: No address/Inst—Patient had no fixed address or currently resides in an institution (e.g., shelter, jail, hospital). For the
data table accompanying this figure, see Appendix Table C3.
The West region accounted for the highest percentage of fentanyl and heroin-related ED visits, while the
South region accounted for the highest percentage of other opioid-related ED visits.
Figure 4.B.6 Fentanyl, heroin, and other opioid-related ED visits by quarter
40%
20%
0%
Qtr1 Qtr2 Qtr3 Qtr4 Qtr1 Qtr2 Qtr3 Qtr4 Qtr1 Qtr2 Qtr3 Qtr4
Fentanyl-related ED visits rose throughout 2021, peaking in quarter 4. Heroin-related ED visits rose from quarter 1 through
quarter 3, and declined in quarter 4. Other opioid-related ED visits peaked in quarter 2, and declined in quarters 3 and 4.

Key Takeaways from Opioid-Related ED Visits

Fentanyl, heroin, and other opioid-related ED visits in 2021 were more likely to be Not Hispanic or Latino, White,
male, and among those aged 26 to 44. While males accounted for the majority of fentanyl, heroin, and other opioid-
related ED visits, the difference between males and females was smaller for other opioid-related ED visits (54.04%
and 45.94% respectively). Black or African American patients accounted for 20.35 percent of all heroin-related ED
visits, but a smaller portion for fentanyl (8.39%) and other opioid-related ED visits (10.59%). Other opioid-related
ED visits peaked in quarter 2, heroin-related ED visits peaked in quarter 3, and fentanyl-related ED visits peaked in
quarter 4. For data tables accompanying the figures in this section, see Appendix Table C3.

4.C Estimated Percentages of Methamphetamine-Related ED Visits
(Estimated weighted count = 797,936)
Figure 4.C.1 Methamphetamine-related ED visits by age Figure 4.C.2 Methamphetamine-related ED visits by sex

group
80% 80%
60% 60%
40% 40%
20% 20%
0% 0%
Under 18 18 to 25 26 to 44 45 to 64 65 and over Missing* Male Female Other* Not documented*
The percentage of methamphetamine-related ED visits was Male patients accounted for 69.10 percent of methamphetamine-
highest among patients ages 26 to 44 (59.58%) followed by related ED visits, while female patients accounted for 30.76
patients ages 45 to 64 (28.05%). percent of these visits.
Figure 4.C.3 Methamphetamine-related ED visits by race Figure 4.C.4 Methamphetamine-related ED visits by

ethnicity
100% 100%
80% 80%
60% 60%
40% 40%
20% 20%
0% 0%
White Black Other Multi-Racial Asian AI/AN NH/PI Missing/ND* Not Hispanic or Latino Hispanic or Latino Unknown*
White patients accounted for the highest percentage of Not Hispanic or Latino patients accounted for 67.94 percent of
methamphetamine-related ED visits (62.45%), while Black methamphetamine-related ED visits, while Hispanic or Latino
patients accounted for the second highest percentage of these patients accounted for 12.92 percent of these visits.
visits (9.95%).
medical record.

Estimated Percentages of Methamphetamine-Related ED Visits (continued)
Figure 4.C.5 Methamphetamine-related ED visits by Figure 4.C.6 Methamphetamine-related ED visits by

census region quarter
80% 80%
60% 60%
40% 40%
20% 20%
0% 0%
South West No address/Inst Midwest Missing Northeast* Qtr1 Qtr2 Qtr3 Qtr4
* Suppressed due to a relative standard error (RSE) > 0.5
The regional percentage of methamphetamine-related ED visits The percentage of methamphetamine-related ED visits rose from
was highest among patients residing in the South (35.69%), quarter 1 to quarter 2 and then slightly declined in quarters 3 and 4.
followed closely by patients residing in the West (35.33%).
Note: No address/Inst—Patient had no fixed address or currently resides
in an institution (e.g., shelter, jail, hospital).
Key Takeaways from Methamphetamine-Related ED Visits

Methamphetamine-related ED visits in 2021 were more likely to be Not Hispanic or Latino, White, male, among
those aged 26 to 44, and from the South census region. White patients accounted for the majority (62.45%) of all
methamphetamine-related ED visits, followed by Black or African American patients (9.95%). The South and West
census regions accounted for over 70 percent of methamphetamine-related ED visits. Methamphetamine-related ED
visits peaked in quarter 2 before declining in quarters 3 and 4. For the data table accompanying the figures in this
section, see Appendix Table C4.

4.D Estimated Percentages of Marijuana-Related ED Visits
Figure 4.D.1 Marijuana-related ED Visits by Age Group Figure 4.D.2 Marijuana-related ED Visits by Sex
80% 80%
60% 60%
40% 40%
20% 20%
0% 0%
Under 18 18 to 25 26 to 44 45 to 64 65 and over Missing* Male Female Other* Not documented*
The percentage of marijuana-related ED visits was highest among Male patients accounted for 61.20 percent of marijuana-related
patients ages 26 to 44 (45.20%) followed by patients ages 18 ED visits, while female patients accounted for 38.76 percent of
to 25 (26.80%). these visits.
Figure 4.D.3 Marijuana-related ED visits by race Figure 4.D.4 Marijuana-related ED visits by ethnicity
100% 100%
80% 80%
60% 60%
40% 40%
20% 20%
0% 0%
White Black Other Asian Multi-Racial AI/AN NH/PI Missing/ND* Not Hispanic or Latino Hispanic or Latino Unknown*
White patients accounted for the highest percentage of marijuana- Not Hispanic or Latino patients accounted for 72.65 percent of
related ED visits (50.23%), while Black patients accounted for the marijuana-related ED visits, while Hispanic or Latino patients
second highest percentage of these visits (23.87%). accounted for 12.95 percent of these visits.
medical record.

Estimated Percentages of Marijuana-Related ED Visits (continued)
Figure 4.D.5 Marijuana-related ED visits by census region Figure 4.D.6 Marijuana-related ED visits by quarter
80% 80%
60% 60%
40% 40%
20% 20%
0% 0%
South West Midwest Northeast No address/Inst Missing Qtr1 Qtr2 Qtr3 Qtr4
The regional percentage of marijuana-related ED visits was The percentage of marijuana-related ED visits rose from quarter
highest among patients residing in the South (26.87%), followed 1 to quarter 2 before declining in quarters 3 and 4.
by patients residing in the West (25.04%) and Midwest (22.30%).
Note: No address/Inst—Patient had no fixed address or currently resides

in an institution (e.g., shelter, jail, hospital).
Key Takeaways from Marijuana-Related ED Visits

Marijuana-related ED visits in 2021 were more likely to be Not Hispanic or Latino, White, male, and among those
aged 26 to 44. Patients aged 18-25 accounted for the second highest percentage of these ED visits. While marijuana-
related ED visits were highest in the South census region, each census region accounted for at least 20 percent of
these visits. Marijuana-related ED visits peaked in quarter 2 before declining in quarters 3 and 4. For the data table
accompanying the figures in this section, see Appendix Table C5.

4.E Estimated Percentages of Cocaine-Related ED Visits
Figure 4.E.1 Cocaine-related ED Visits by Age Group Figure 4.E.2 Cocaine-related ED Visits by Sex
80% 80%
60% 60%
40% 40%
20% 20%
0% 0%
Under 18 18 to 25 26 to 44 45 to 64 65 and over Missing* Male Female Other* Not documented
The percentage of cocaine-related ED visits was highest among Male patients accounted for 71.71 percent of cocaine-related
patients ages 26 to 44 (45.52%) followed by patients ages 45 ED visits, while female patients accounted for 28.20 percent
to 64 (38.60%). of these visits.
Figure 4.E.3 Cocaine-related ED visits by race Figure 4.E.4 Cocaine-related ED visits by ethnicity
100% 100%
80% 80%
60% 60%
40% 40%
20% 20%
0% 0%
Black White Other Asian Multi-Racial AI/AN NH/PI Missing/ND* Not Hispanic or Latino Hispanic or Latino Unknown
Black patients accounted for the highest percentage of cocaine- Not Hispanic or Latino patients accounted for 80.74 percent
related ED visits (44.19%), while White patients accounted for of cocaine-related ED visits, while Hispanic or Latino patients
the second highest percentage of these visits (35.11%). accounted for 13.10 percent of these visits.
medical record.

Estimated Percentages of Cocaine-Related ED Visits (continued)
Figure 4.E.5 Cocaine-related ED visits by census region Figure 4.E.6 Cocaine-related ED visits by quarter
80% 80%
60% 60%
40% 40%
20% 20%
0% 0%
Northeast West No address/Inst Missing Midwest* South* Qtr1 Qtr2 Qtr3 Qtr4
The regional percentage of cocaine-related ED visits was highest The percentage of cocaine-related ED visits rose from quarter
among patients residing in the Northeast (26.84%), followed 1 to quarter 2 and then declined in quarters 3 and 4.
by patients residing in the West (13.61%).
Note: No address/Inst—Patient had no fixed address or currently resides in
an institution (e.g., shelter, jail, hospital).
Key Takeaways from Cocaine-Related ED Visits

Cocaine-related ED visits in 2021 were more likely to be Not Hispanic or Latino, Black or African American, male,
and among those aged 26 to 44. Patients ages 45-64 accounted for the second highest percentage of these visits.
Cocaine-related ED visits peaked in quarter 2 before declining in quarters 3 and 4. For the data table accompanying
the figures in this section, see Appendix Table C6.

5. Key Findings from Sentinel Hospitals, April 2021 – November
2021
This section presents unweighted data from eight sentinel hospitals that submitted complete data from April 2021 to
November 2021. The top ten drugs involved in drug-related ED visits are shown in Table 3. This analysis was conducted
using drug IDs from the DRV.
5.A Sentinel Hospitals: Top Ten Drugs in Drug-Related ED Visits
In sentinel hospitals from April to November 2021, alcohol-related ED visits accounted for more than half of all drug-
related ED visits (range 49.91-52.48%), followed by methamphetamine (range 16.39-18.56%), and cocaine (range
8.16-9.34%). About 1.74 to 3.77 percent of drug-related ED visits involved drugs unknown to the patient. When
examining the percentages of monthly drug-related ED visits from April to November 2021, ED visits involving alcohol,
methamphetamine, marijuana, cocaine, and heroin decreased, whereas ED visits involving narcotic analgesic-NOS
(not otherwise specified) and fentanyl increased.
Table 5.A.1 Percentage of top ten drug-related visits to all drug-related ED visits by month
Apr-21 May-21 Jun-21 Jul-21 Aug-21 Sep-21 Oct-21 Nov-21

Drug name
(n=6,822) (n=6,620) (n=6,327) (n=6,072) (n=5,513) (n=5,409) (n=5,476) (n=5,413)
Alcohol
52.48 52.24 49.91 51.76 51.10 51.99 51.75 50.75
(N = 24,549)
Methamphetamine
18.56 16.87 17.70 17.21 17.56 17.58 16.65 16.39
(N = 8,266)
Marijuana
11.92 9.47 9.36 9.80 9.70 10.22 10.99 10.51
(N = 4,886)
Cocaine
9.34 8.16 8.42 8.93 8.83 9.11 9.26 8.48
(N = 4,198)
Heroin
7.04 5.82 6.34 5.90 5.73 5.58 4.69 4.75
(N = 2756)
Fentanyl
4.35 4.21 4.90 5.58 5.48 5.86 6.15 5.32
(N = 2,469)
Narcotic analgesics-NOS1
2.54 3.08 3.27 3.33 3.23 4.03 3.51 3.55
(N = 1,566)
Drug unknown
1.74 2.95 2.96 3.18 3.36 3.33 3.21 3.77
(N = 1,439)
Amphetamine
1.39 1.18 1.23 1.68 1.27 1.09 1.26 1.50
(N = 632)
Alprazolam
1.23 1.40 1.63 0.99 1.38 1.44 0.91 1.29
(N = 614)
n = number of total drug-related ED visits for the month

N = total number of drug-related ED visits for April 2021 – November 2021
1
NOS = not otherwise specified
Note: The above table presents unduplicated counts by drug ID. An ED visit that involves two brands linked to the same drug ID, for example Benzedrine
and Obetrol, will be counted only once as Amphetamine.

5.B Sentinel Hospitals: Polysubstance ED Visits
DAWN defines polysubstance ED visits as visits related to more than one drug. For example, if cocaine and marijuana
were both documented as related to an ED visit, it is considered a polysubstance visit. The top six drugs involved in
polysubstance visits in sentinel hospitals from April to November 2021 were alcohol, methamphetamine, marijuana,
cocaine, heroin, and fentanyl (Figure 5.B.1). While a majority of alcohol-related ED visits involved alcohol only, a
significant proportion of methamphetamine-, marijuana-, cocaine-, heroin-, and fentanyl-related ED visits involved
at least one other drug.
Figure 5.B.1 Polysubstance ED visits
30,000
20,000
COUNT
10,000
Alcohol Methamphetamine Marijuana Cocaine Heroin Fentanyl

Polysubstance 5,353 4,027 3,106 2,722 1,560 1,539
Drug only 19,196 4,239 1,780 1,476 1,196 930
Note: DAWN defines polysubstance ED visits as (1) ED visits involving more than one known and/or unknown drug irrespective of whether they are taken
together or not, (2) ED visits involving unknown polydrugs, and (3) ED visits involving a combination of pharmaceutical substance with an illicit drug.
Polysubstance ED visits for the top six substances were analyzed further to identify the most common additional drug
involved in those ED visits (Figures 5.B.2 – 5.B.7). Alcohol was the most common drug involved in methamphetamine-
(35.36%), marijuana- (52.74%), and cocaine- (53.20%) related polysubstance ED visits, while methamphetamine was the
most common drug involved in heroin- (48.65%) and fentanyl- (45.81%) related polysubstance ED visits. The N in the title
of figures 5.B.2 through 5.B.7 represents the total number of polysubstance ED visits for each highlighted drug.

Figure 5.B.2 Top five drugs involved in alcohol-related Figure 5.B.3 Top five drugs involved in methamphet-
polysubstance ED visits (N = 5,353) amine-related polysubstance ED visits (N = 4,027)
60% 60%
40% 40%
35.36
30.60
27.05 26.60
23.94
20% 20% 18.85
17.58 17.51
6.95
5.40
0 0
Marijuana Cocaine Methamphetamine Heroin Drug unknown Alcohol Marijuana Heroin Cocaine Fentanyl
Figure 5.B.4 Top five drugs involved in marijuana-related Figure 5.B.5 Top five drugs involved in cocaine-related
polysubstance ED visits (N = 3,106) polysubstance ED visits (N = 2,722)
60% 60%
53.20
52.74
40% 40%
31.04 28.62
26.01
25.08
20% 20%
12.27
8.89
7.21
5.51
0 0
Alcohol Methamphetamine Cocaine Fentanyl Amphetamine Alcohol Marijuana Methamphetamine Heroin Fentanyl
Figure 5.B.6 Top five drugs involved in heroin-related Figure 5.B.7 Top five drugs involved in fentanyl-related
polysubstance ED visits (N = 1,560) polysubstance ED visits (N = 1,539)
60% 60%
48.65 45.81
40% 40%
23.85 21.44
21.41 21.15 18.00
20% 20%
15.72 14.55
8.97
0 0
Methamphetamine Alcohol Cocaine Fentanyl Marijuana Methamphetamine Heroin Alcohol Cocaine Marijuana
Note: Percentages presented in Figures 5.B.2 – 5.B.7 above are not mutually exclusive and may add up to more than 100 percent. ED visits involving more
than one drug are counted once for each drug. For example, an alcohol-related polysubstance ED visit that involves both marijuana and cocaine will be
counted in marijuana as well as cocaine.

6. Drugs New to DAWN Drug Reference Vocabulary, 2021
When a drug involved in an ED visit is not listed in the DAWN DRV, it is recorded verbatim from the ED record. These
new drugs are reviewed regularly for inclusion and are added to the DAWN DRV upon SAMHSA approval. In 2021, there
were 38 substances new to the DAWN DRV and, after research as to their make-up, all 38 were added to the DRV and
assigned to existing categories.
Of the 38, 11 (28.94%) were illicit drugs, and 27 (71.05%) were non-illicit substances. The majority of the new illicit drugs were
stimulants (n=5) followed by cannabinoids (n=3). The largest groups of non-illicit substances were herbal products (n=6),
followed by CNS stimulants (n=3). Table 6.1 presents the complete list of 38 substances new to the DAWN DRV in 2021.
Table 6.1 Drugs new to the DAWN DRV, 2021
Category DAWN DRV category

Drugs new to DAWN DRV in 2021
(number identified) (number identified)
Illicit drugs (11)
Bubble, MADAM-6, Meth Laced Cigar, Synthetic

Stimulants (5)
Amphetamine, Synthetic Cocaine
Cannabinoids (3) Candy Cane, Moon Rocks, Trees (marijuana)
Cocaine (1) Sniff
MDMA (Ecstasy) (1) Sassafras (MDMA)
Miscellaneous hallucinogens (1) Mushroom Bar
Non-Illicit substances (27)
EZ Sleep, Harmala/Syrian Rue, Legal Lean, Lipo-Gel,

Herbal products (6)
Mugwort, Zarbees
Antidepressants (2) Tianaa, Za Za
CNS stimulants (3) Adrafinil, Cyclazodone, Phenethylamine
Dermatological agents (2) Boric Acid Suppository, Eczema Cream
Erectile dysfunction agents (2) Ant Male Enhancement, Royal Honey VIP
Nutraceutical products (2) Dherbs Full Body Cleanse, Triple Action Joint Health
Oral nutritional supplements (2) Animal Cuts, Arachidonic Acid
Anticonvulsants (1) CBD
Antihistamines (1) Cinnarizine
CD20 monoclonal antibodies (1) Ublituximab
Immune globulins (1) Brillia
Miscellaneous antineoplastics (1) Magrolimab
Sex hormones (1) Ingesta
Smoking cessation agents (1) ZYN Nicotine Pouches
Vitamins (1) Vitamin B5

Appendix A. Methods
DAWN Hospital Selection Design
The DAWN sample of 53 hospitals was created from a hybrid design that includes a group of ten hospitals identified
for sentinel surveillance (Part A), three early sampled hospitals considered in Part A for weighting and estimation, a
probability sample of eight hospitals from high priority rural and suburban counties (Part B), and a probability sample
of 32 hospitals from all remaining counties (Part C). Hospitals in the probability sample-based surveillance were selected
using a stratified random sampling method. The 2016 American Hospital Association (AHA) Survey database1 was used
to identify the universe of eligible hospitals. Two datasets were used to identify high priority counties: (1) the Centers
for Disease Control and Prevention’s Wide-Ranging Online Database for Epidemiologic Research (CDC WONDER)2
for the 2012-2016 county-level rates of opioid, cocaine, and stimulant-related mortality (Part A) or the highest rates of
5-year combined drug-related overdose deaths (Parts B and C) and (2) the Agency for Healthcare Research and Quality’s
(AHRQ) Healthcare Cost Utilization Project (HCUP)3 for county-level counts and rates of morbidity (Part A). County-level
urbanization was derived from the National Center for Health Statistics’ (NCHS) 2013 Urban-Rural Classification Scheme.4
PART A
Part A consists of ten sentinel hospitals. This is a set of large urban hospitals from high-priority sentinel areas, defined
as counties with high counts and rates of opioid, cocaine, and stimulant-related mortality and high counts and rates of
morbidity. Geographic distribution and characteristics of interest were also considered, for example, statewide legalization
of marijuana or early incidence of novel psychoactive substances. Three early sampled hospitals were added to Part A
during a sample redesign in 2019. They are not identified as sentinel hospitals, their data is not used in sentinel hospital
analysis, but they are included in Part A of the sample design for weighted estimates.
PART B
Part B is a systematic random sample of eight rural and suburban hospitals located in high-priority counties. The top rural
and suburban counties rates of 5-year combined drug-related overdose deaths were ranked highest to lowest for sampling.
Hospitals in Part B were stratified by 1 stratum with 4 levels: U.S. census regions Northeast, South, Midwest, and West.
Two hospitals were selected per stratum.
PART C
Part C is a systematic random sample of 32 hospitals from counties not included in Part B. The hospitals in Part C were
stratified by 3 strata with 16 levels total: U.S. census regions (4 levels) and their counties drug-induced 5-year combined
death rate (2 levels) and emergency department annual visit counts (2 levels) were ranked highest to lowest for sampling.
Two hospitals were selected per stratum.
1
https://www.ahadata.com/aha-annual-survey-database-asdb/
2
https://wonder.cdc.gov/mcd-icd10.html
3
https://www.hcup-us.ahrq.gov/overview.jsp
4
https://www.cdc.gov/nchs/data/series/sr_02/sr02_166.pdf. Six-level urbanization categories from NCHS were collapsed to generate three
urbanization categories (urban, suburban, and rural) in DAWN.

Table A1 provides an overview of the DAWN hospitals by Part and Strata, and levels within each strata.
Table A1 Overview of the DAWN sample
DAWN Number of
Strata (H) and Levels (L) Brief description
sample hospitals (n)
Consists of ten sentinel hospitals and three
Part A N/A 13
hospitals recruited prior to redesign.
Systematic random sample of two hospitals

Census region (H): Northeast, South,
Part B per stratum from suburban and rural counties 8
Midwest, and West (L = 4)
with a high rate of drug-induced deaths.
• Census region (L = 4)
Systematic random sample of two hospitals
• High vs. Low Drug-Induced Death Counts
Part C per stratum excluding hospitals included in 32
for Hospital County (L = 2)
Part B.
• High vs. Low AHA ED Visit Counts (L = 2)

Figure A1 provides an overview of the DAWN hospital selection design.
Figure A1 DAWN hospital selection design
DAWN Hospitals
(n = 53)
The hybrid design includes sentinel-based surveillance,

and sample-based surveillance from high priority rural
and suburban counties and all remaining counties.
Part A: Part B: Part C:
Sentinel Sample from High Sample from

Hospitals Priority Rural All Remaining
and Suburban Counties
Counties
Strata: Strata:
Regions (4): Regions (4)
Northeast, South,
High/Low drug-
Midwest, and West
induced death
count (2)
High/Low ED visit
count (2)
Method: Method: Method:

10 large hospitals Systematic Systematic
from key urban random sample random sample of
areas and 3 of two hospitals two hospitals per
previously per stratum from stratum excluding
sampled hospitals the top rural counties included
and suburban in Part B
countries with the
highest drug-
induced death rate
Sample size = 13 Sample size = 8 Sample size = 32

Data Abstraction
DAWN medical record abstractors are trained on how to thoroughly review ED records, identify drug-related visits
following the DAWN criteria, and abstract DAWN data elements. A drug-related visit is defined as an ED visit related to
or induced by recent drug use.
Data elements are abstracted from an ED visit if the patient was treated in the ED for a condition that was induced by
or related to recent use of a drug. ED visits when the patient left the ED without being seen by a clinician, or visits when
the patient came to the ED but was directly admitted to the hospital, are ineligible for review.
Data elements abstracted for DAWN include patient visit characteristics (date/time of visit), patient demographics
(age, sex, race, ethnicity, county of residence), drugs involved, ED diagnoses, patient disposition, and a brief narrative
describing how the drug was involved in the visit.
Data Quality Review
Prior to the weighting and estimation effort, a quality review was conducted on the preliminary 2021 data. Checks were
run on date of visit, and data presence in 2021, by year, quarter, and month. Analytic tests are used to identify out-of-
control points by month in ED visits, records reviewed, and count of drug-related visits. Out-of-control points from the
analysis were identified and investigated by survey operations to confirm or resolve identified issues.
Since DAWN reviews every ED visit from participating hospitals, there was no sampling error within a hospital. Missing
data was minimal, which reduced the potential for bias beyond measurement error and/or data quality.
Weighting and Estimation
Given DAWN’s hybrid sentinel surveillance and probability sample design, DAWN employed a multi-step weighting
process to produce nationally representative estimates. The multi-step weighting process involved (1) calculating initial
base weights for each sampling part/stratum, (2) adjusting the initial base weights for changes in the sample design and
sampling frame, (3) adjusting for hospital non-response, and (4) post-stratification to adjust DAWN estimates of ED visit
totals to AHA ED visits for the given stratum.
STEP 1:
Initial base weights for each stratum were generated using the inverse of the selection probability. Since part A hospitals
were chosen by SAMHSA with certainty, their initial base weight was set to 1. For Part B and Part C, base weights for
sampled hospitals in each stratum was set to NH/2, where NH is the number of hospitals in the sampling frame in the
given explicit stratum (H). Table A2 below provides NH and base weights for each stratum for Parts B and C.

Table A2 Initial base weights for Part B and Part C hospitals
Number of hospitals on
Part Stratum Initial base weight
the sampling frame (NH)
B Midwest 19 9.5
B Northeast 27 13.5
B South 20 10
B West 26 13
C MHH 292 146
C MHL 70 35
C MLH 499 249.5
C MLL 433 216.5
C NHH 183 91.5
C NHL 44 22
C NLH 190 95
C NLL 87 43.5
C SHH 492 246
C SHL 54 27
C SLH 654 327
C SLL 408 204
C WHH 226 113
C WHL 83 41.5
C WLH 290 145
C WLL 215 107.5
STEP 2:
Initial base weights were adjusted to reflect changes to DAWN sample design and sampling frame, and address duplicate
records from the sampling frame. Changes to DAWN sample design (i.e., DAWN sample redesign) resulted in moving
three initially sampled hospitals to Part A. Hospitals with duplicate records and those with less than 100 annual visits
were removed from the sampling frame. As a result, the base weights of the Part B and Part C hospitals were adjusted by
a factor of N*H/NH, where N*H is the updated count of hospitals in a given stratum (H) after removing Part A hospitals,
duplicates, and hospitals with less than 100 ED visits. No adjustment was required to the Part A hospitals, i.e., beyond the
affect they had on the base weights for other hospitals in Parts B and C. Table A3 provides updated NH and adjusted base
weights for each stratum for Parts B and C.

Table A3 Adjusted base weights for Part B and Part C hospitals5
Part Stratum Updated count of hospitals (N*H) Adjusted base weight

B Midwest 19 9.5
B Northeast 27 13.5
B South 20 10
B West 25 12.5
C MHH 290 145
C MHL 70 35
C MLH 499 249.5
C MLL 432 216
C NHH 183 91.5
C NHL 44 22
C NLH 190 95
C NLL 87 43.5
C SHH 488 244
C SHL 54 27
C SLH 647 323.5
C SLL 407 203.5
C WHH 221 110.5
C WHL 83 41.5
C WLH 290 145
C WLL 212 106
STEP 3:
The adjusted base weights were used and further adjusted to generate quarterly nonresponse-adjusted weights. Depending
on data availability by hospital in a given quarter, DAWN employed two types of nonresponse adjustment. For hospitals
with at least one month of completed data in the quarter, quarterly weights were adjusted by the ratio of the number
of days in the quarter to the number of non-missing days in the quarter. For hospitals with no data in the quarter, the
weights of other hospitals in the same or similar stratum, after collapsing, were adjusted to represent both the responding
and nonresponding hospitals in that collapsed stratum.
STEP 4:
The next step in the weighting process was to calibrate (poststratify) the estimated number of annual ED visits from the
responding DAWN 2021 hospitals to the total number of ED visits in DAWN-eligible hospitals from the 2020 AHA Annual
Survey Database for Parts B and C. The poststratification adjustment was done within collapsed strata created in Step 3.
Within a poststratification stratum, the adjustment factor is equal to the ratio of the AHA ED visit total to the weighted
DAWN estimate of ED visits.
STEP 5:
The final step required creation of the final weight variable for generating 2021 DAWN estimates, calculated as:
Final Weight = Initial Base Weight*Base Weight Adjustment Factor*Nonresponse Adjustment Factor*Poststratification Factor.
5
There were 4,288 records total = 4,312 from Table A2, less: (A) The 13 hospitals moved to Part A, (B) the nine duplicate hospitals, and (C) the two
hospitals in the two strata where the sample was not reselected to reflect the change in the target population definition to exclude hospitals with less
than 100 EDVs. Note that the seven hospitals in stratum C-SHL are already excluded from the 4,312 records in Table A2.

Appendix B. Definitions of the Top Five Drugs in Drug-Related
ED Visits
Table B1 Top five drug definitions
Top five drugs Drugs included

Alcohol-related ED visits alcohol (ethanol) and alcohol/food products
heroin, fentanyl/heroin, cocaine/heroin, heroin/methamphetamine, heroin/

Opioids: heroin-related ED visits methylenedioxymethamphetamine, heroin/marijuana, alprazolam/heroin, amphetamine/
heroin, cyanide/heroin/strychnine, heroin/benzodiazepine, heroin/desomorphine
fentanyl, fentanyl/heroin, fentanyl/percocet, fentanyl/methamphetamine, cocaine/

Opioids: fentanyl-related ED visits fentanyl, fentanyl/marijuana, fentanyl/oxycodone, alprazolam/fentanyl, carfentanil,
acetaminophen/fentanyl, fentanyl/benzodiazepine, fentanyl/methadone
narcotic analgesics-NOS, acetaminophen-oxycodone, oxycodone, methadone,

buprenorphine-naloxone, acetaminophen-hydrocodone, tramadol, morphine,
hydrocodone, buprenorphine, hydromorphone, acetaminophen-codeine, codeine,
oxymorphone, opium, codeine combination-NOS, marijuana/opioid, cocaine/
Opioids: other opioid-related ED visits opioids, methamphetamine/opioids, cocaine/marijuana/opioid, desomorphine,
hydrocodone-ibuprofen, acetaminophen-oxycodone/alprazolam, heroin/desomorphine,
acetaminophen-oxycodone/marijuana, acetaminophen/butalbital/caffeine/codeine,
morphine/nicotine, pentazocine, propoxyphene, acetaminophen-tramadol, butorphanol,
tapentadol
methamphetamine, fentanyl/methamphetamine, marijuana/methamphetamine, heroin/

methamphetamine, methamphetamine/opioids, cocaine/methamphetamine, heroin/
Methamphetamine-related ED visits methylenedioxymethamphetamine, cocaine/methylenedioxymethamphetamine,
methamphetamine/methylenedioxymethamphetamine, cocaine/marijuana/
methamphetamine
marijuana, marijuana/methamphetamine, fentanyl/marijuana, cocaine/marijuana,

marijuana/opioid, marijuana/phencyclidine, cocaine/marijuana/opioid, heroin/marijuana,
Marijuana-related ED visits marijuana/synthetic cannabinoid, amphetamine/marijuana, marijuana/lysergic acid
diethylamide, cocaine/marijuana/methamphetamine, acetaminophen-oxycodone/
marijuana
cocaine, cocaine/heroin, cocaine/fentanyl, cocaine/marijuana, cocaine/opioids, cocaine/

Cocaine-related ED visits methamphetamine, cocaine/marijuana/opioid, amphetamine/cocaine, cocaine/
methylenedioxymethamphetamine, cocaine/marijuana/methamphetamine

Appendix C. Data Tables, Top Five Drugs in Drug-Related ED
Visits, 2021
Table C1 Estimated percentages and counts of drug-related ED visits by top five drugs, 2021
Percent
Top five drugs* 95% CI
(estimated counts)
39.33%
Alcohol 33.40 – 45.27
(N = 2,942,609)
14.07%
Opioids1 10.05 – 18.09
(N = 1,032,510)
1.76%
Fentanyl 0.83 – 2.69
(N = 122,884)
6.94%
Heroin 3.20 – 10.67
(N = 493,916)
6.77%
Other opioids 5.51 – 8.03
(N = 484,735)
11.02%
Methamphetamine 7.69 – 14.36
(N = 797,936)
10.78%
Marijuana 8.84 – 12.71
(N = 787,769)
4.71%
Cocaine 2.41 – 7.01
(N = 336,516)
* A significant number of ED visits involve multiple substances, therefore, caution should be exercised
while adding and/or comparing estimates across drugs, as ED visits related to more than one drug
are counted more than once. For example, an ED visit related to both cocaine and marijuana are
counted in both cocaine-related and marijuana-related ED visits.
1
The opioids category was further broken down to Fentanyl, Heroin, and Other Opioids. However,
since patients taking opioids that fall under different opioid categories (e.g. Fentanyl and Other
Opioids) are counted in each category, the total ED visits for Fentanyl, Heroin, and Other Opioids
exceeds the Opioids ED visits.

Table C2 Estimated subgroup percentages of alcohol-related ED visits (N = 2,942,609)
Subgroups Percent 95% CI

Age groups
Under 18 1.18 0.77 – 1.59
18 to 25 9.99 9.17 – 10.80
26 to 44 40.63 37.93 – 43.33
45 to 64 38.92 36.71 – 41.12
65 and over 8.77 7.86 – 9.68
Missing * *
Age groups (< 21 and 21+)
Under 21 3.90 3.23 – 4.57
21 and over 95.58 94.51 – 96.65
Missing * *
Sex
Male 71.01 68.94 – 73.09
Female 28.94 26.82 – 31.05
Other * *
Not documented 0.01 0.00 – 0.01
Race
White 59.76 42.60 – 76.92
Black or African American 14.12 4.90 – 23.33
Asian 1.29 0.58 – 2.00
American Indian/Alaska Native * *
Native Hawaiian/Pacific Islander 0.24 0.19 – 0.29
Other race1 15.09 7.50 – 22.67
Multi-Racial 2
0.71 0.25 – 1.17
Missing/Not documented * *
Ethnicity
Not Hispanic or Latino 74.65 63.92 – 85.38
Hispanic or Latino 14.26 10.71 – 17.81
Unknown 11.09 0.00 – 22.28
Region
Northeast 23.42 7.52 – 39.31
Midwest 22.40 7.68 – 37.13
South 16.39 9.02 – 23.75
West 27.45 10.17 – 44.74
No address/Institution3 8.15 3.49 – 12.82
Missing 2.18 1.09 – 3.27
Quarter
Quarter 1 20.82 19.55 – 22.10
Quarter 2 28.02 26.21 – 29.82
Quarter 3 27.35 26.45 – 28.26
Quarter 4 23.81 22.40 – 25.22
* Estimates with relative standard error (RSE) > 0.5 are suppressed.
1
Other race—The race documented in the medical record does not fit any other race category.
2
ED visits with multiple-race categories are counted in the Multi-Racial category only.
3
No address/Institution—Patient had no fixed address or currently resides in an institution
(e.g., shelter, jail, hospital).

Table C3 Estimated subgroup percentages of fentanyl, heroin, and other opioid-related ED visits
Fentanyl-related ED visits Heroin-related ED visits Other opioid-related ED visits

(N = 122,884) (N = 493,916) (N = 484,735)
Subgroups Percent 95% CI Percent 95% CI Percent 95% CI

Age groups
Under 18 * * 0.39 0.08 – 0.70 2.84 1.37 – 4.31
18 to 25 16.90 13.16 – 20.64 7.95 5.01 – 10.89 8.95 6.74 – 11.17
26 to 44 62.76 55.18 – 70.33 57.56 46.44 – 68.68 44.84 39.66 – 50.01
45 to 64 14.65 11.34 – 17.97 30.10 20.04 – 40.15 29.07 24.97 – 33.17
65 and over 3.67 1.07 – 6.26 3.56 0.52 – 6.59 14.05 10.07 – 18.04
Missing * * * * * *
Sex
Male 66.35 58.43 – 74.27 70.47 64.87 – 76.07 54.04 47.51 – 60.57
Female 33.64 25.72 – 41.56 29.52 23.90 – 35.13 45.94 39.39 – 52.48
Other — — — — — —
Not documented — — — — — —
Race
White 67.74 55.00 – 80.48 61.26 36.00 – 86.53 67.37 47.00 – 87.74
Black or African American 8.39 4.54 – 12.24 20.35 3.83 – 36.86 10.59 3.36 – 17.82
Asian 1.13 0.05 – 2.20 * * 0.85 0.51 – 1.19
American Indian/Alaska Native 0.34 0.24 – 0.45 0.23 0.00 – 0.46 0.59 0.27 – 0.92
Native Hawaiian/
* * * * 0.11 0.05 – 0.17
Pacific Islander
Other race1 12.44 9.16 – 15.73 11.34 2.92 – 19.76 8.65 3.78 – 13.51
Multi-Racial 2
1.39 0.41 – 2.36 * * * *
Missing/Not documented * * * * * *
Ethnicity
Not Hispanic or Latino 70.83 59.82 – 81.85 79.82 71.67 – 87.97 75.64 59.14 – 92.13
Hispanic or Latino 15.34 5.68 – 25.00 8.99 4.41 – 13.56 9.25 5.83 – 12.67
Unknown 13.83 0.99 – 26.67 11.19 2.76 – 19.63 * *
Region
Northeast 16.79 7.76 – 25.82 * * 15.27 6.77 – 23.78
Midwest 13.44 5.06 – 21.82 * * 19.57 4.66 – 34.48
South 16.69 5.51 – 27.87 * * 33.48 16.41 – 50.55
West 39.65 25.50 – 53.80 13.00 0.78 – 25.22 26.03 11.47 – 40.59
No address/Institution3 12.22 2.70 – 21.74 6.22 1.25 – 11.19 3.89 2.33 – 5.45
Missing 1.22 0.72 – 1.71 1.97 0.38 – 3.55 1.75 0.47 – 3.02
Quarter
Quarter 1 15.99 10.87 – 21.11 20.93 16.09 – 25.77 19.56 16.95 – 22.18
Quarter 2 26.34 22.55 – 30.12 27.25 24.83 – 29.66 31.09 29.28 – 32.9
Quarter 3 28.71 22.98 – 34.45 28.45 24.65 – 32.26 26.00 23.89 – 28.12
Quarter 4 28.96 24.40 – 33.51 23.37 21.71 – 25.03 23.34 21.28 – 25.40
— Estimates based on less than 10 cases were suppressed to ensure patient confidentiality.
1
2
3
No address/Institution—Patient had no fixed address or currently resides in an institution (e.g., shelter, jail, hospital).

Table C4 Estimated subgroup percentages of methamphetamine-related ED visits (N = 797,936)

Age groups
Under 18 1.04 0.53 – 1.56
18 to 25 9.25 7.74 – 10.77
26 to 44 59.58 56.63 – 62.52
45 to 64 28.05 25.29 – 30.80
65 and over 2.00 0.94 – 3.05
Missing * *
Sex
Male 69.10 66.78 – 71.42
Female 30.76 28.41 – 33.11
Other * *
Not documented * *
Race
White 62.45 37.88 – 87.02
Asian 1.03 0.40 – 1.67
American Indian/Alaska Native 0.54 0.12 – 0.96
Other race1 9.84 5.86 – 13.81
Multi-Racial2 1.66 0.09 – 3.23
Ethnicity
Unknown * *
Region
Northeast * *
Midwest 6.63 2.76 – 10.51
South 35.68 15.74 – 55.62
West 35.33 20.94 – 49.72
No address/Institution3 12.45 7.61 – 17.29
Missing 4.02 0.78 – 7.27
Quarter
Quarter 1 21.98 17.93 – 26.04
Quarter 2 28.45 25.75 – 31.15
Quarter 3 25.81 23.64 – 27.98
Quarter 4 23.75 21.40 – 26.10
1
2
3

Table C5 Estimated subgroup percentages of marijuana-related ED visits (N = 787,769)

Age groups
Under 18 8.86 5.60 – 12.12
18 to 25 26.80 24.37 – 29.24
26 to 44 45.20 42.65 – 47.75
45 to 64 16.42 14.53 – 18.31
65 and over 2.52 1.69 – 3.34
Missing * *
Sex
Male 61.20 58.35 – 64.05
Female 38.76 35.88 – 41.63
Other * *
Not documented * *
Race
White 50.23 33.06 – 67.40
Asian 1.22 0.65 – 1.80
Other race 1
13.20 7.27 – 19.13
Multi-Racial2 0.73 0.42 – 1.04
Ethnicity
Unknown * *
Region
Northeast 20.48 6.08 – 34.87
Midwest 22.30 8.14 – 36.46
South 26.87 15.32 – 38.42
West 25.04 12.24 – 37.83
No address/Institution 3
4.16 1.81 – 6.51
Missing 1.15 0.64 – 1.67
Quarter
Quarter 1 20.41 17.66 – 23.16
Quarter 2 28.95 24.94 – 32.96
Quarter 3 26.41 25.18 – 27.64
Quarter 4 24.23 19.80 – 28.65
1
2
3

Table C6 Estimated subgroup percentages of cocaine-related ED visits (N = 336,516)

Age groups
Under 18 1.23 0.04 – 2.42
18 to 25 9.74 5.20 – 14.27
26 to 44 45.52 38.63 – 52.41
45 to 64 38.60 28.84 – 48.36
65 and over 4.28 2.56 – 5.99
Missing * *
Sex
Male 71.71 67.84 – 75.58
Female 28.20 24.27 – 32.13
Other * *
Not documented 0.01 0.00 – 0.02
Race
White 35.11 20.21 – 50.00
Asian 1.09 0.48 – 1.70
Other race 1
14.88 6.77 – 22.99
Multi-Racial2 0.76 0.34 – 1.19
Ethnicity
Unknown 6.16 0.10 – 12.23
Region
Northeast 26.84 1.93 – 51.75
Midwest * *
South * *
West 13.61 2.80 – 24.41
No address/Institution 3
6.87 1.41 – 12.33
Missing 1.21 0.09 – 2.32
Quarter
Quarter 1 22.03 19.13 – 24.93
Quarter 2 28.13 26.06 – 30.19
Quarter 3 27.29 25.63 – 28.95
Quarter 4 22.55 19.63 – 25.48
1
2
3
No address/Institution—Patient had no fixed address or currently resides in an institution (e.g., shelter, jail, hospital).

Publication No. PEP22-07-03-001
SAMHSA’s mission is to reduce the impact of substance abuse and mental illness on America’s communities.
1-877-SAMHSA-7 (1-877-726-4727) ● 1-800-487-4889 (TDD) ● www.samhsa.gov

Pep22 07 03 001

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Pep22 07 03 001

Uploaded by

Copyright:

Available Formats

Preliminary Findings from

Drug Abuse Warning Network

Public Domain Notice

Electronic Access and Printed Copies

SAMHSA Nondiscrimination Notice

U.S. Department of Health and Human Services

Publication Number PEP22-07-03-001

DAWN Preliminary Findings from Drug-related Emergency Department Visits, 2021

2. Introduction and Background 3

5. Key Findings from Sentinel Hospitals, April 2021 – November 2021 21

6. Drugs New to DAWN Drug Reference Vocabulary, 2021 24

Appendix B. Definitions of the Top Five Drugs in Drug-Related ED Visits 31

Appendix C. Data Tables, Top Five Drugs in Drug-Related ED Visits, 2021 32

DAWN Preliminary Findings from Drug-related Emergency Department Visits, 2021 i

Table 3.A.1 Overview of the DAWN hospital selection 4

DAWN Preliminary Findings from Drug-related Emergency Department Visits, 2021 ii

KEY FINDINGS BY DRUGS

KEY FINDINGS BY AGE, SEX, RACE, ETHNICITY, AND CENSUS REGION

1.B Key Findings from Sentinel Hospitals, April 2021 – November 2021

DAWN Preliminary Findings from Drug-related Emergency Department Visits, 2021 1

1.C Drugs New to DAWN’s Drug Reference Vocabulary

DAWN Preliminary Findings from Drug-related Emergency Department Visits, 2021 2

The key objectives of DAWN are to:

DAWN Preliminary Findings from Drug-related Emergency Department Visits, 2021 3

Table 3.A.1 Overview of the DAWN hospital selection

Part Number Description

3.B Data Abstraction, Drug Reference Vocabulary, and Quality Control

DAWN Preliminary Findings from Drug-related Emergency Department Visits, 2021 4

3.C Data Notes and Limitations

3.D Characteristics of Unweighted Drug-Related ED Visits from DAWN Hospitals

DAWN Preliminary Findings from Drug-related Emergency Department Visits, 2021 5

Subgroups Count Percent

Under 18 4,953 3.50

65 and over 13,692 9.67

Missing 559 0.39

Age Groups (< 21 and 21+)

Under 21 9,691 6.85

21 and over 131,279 92.76

Missing 559 0.40

Male 92,920 65.65

Female 48,381 34.18

Other 119 0.08

Not documented 109 0.08

White 79,269 56.01

Black or African-American 33,430 23.62

Asian 3,077 2.17

American Indian/Alaska Native 4,073 2.88

Native Hawaiian/Pacific Islander 1,493 1.05

Other race 13,342 9.43

Multi-Racial* 2,735 1.93

Missing/Not documented 4,110 2.90

Not Hispanic or Latino 110,963 78.40

Hispanic or Latino 19,820 14.00

Unknown 10,746 7.59

Midwest 31,904 22.54

Northeast 23,338 16.49

South 24,768 17.50

West 43,324 30.61

No address/Institution 16,299 11.52

Missing 1,896 1.34

DAWN Preliminary Findings from Drug-related Emergency Department Visits, 2021 6

4. Weighted National Estimates of the Top Five Drugs in Drug-

› Alcohol-related ED visits involving alcohol alone or in combination with other drugs.